Aroma Components of Acid-Hydrolyzed Vegetable ... - ACS Publications

1302 West Pennsylvania Avenue, Urbana, Illinois 61801. Hydrolyzed vegetable protein (HVP) was prepared from rice bran protein concentrate (RBPc) by...
12 downloads 0 Views 58KB Size
3044

J. Agric. Food Chem. 2007, 55, 3044−3050

Aroma Components of Acid-Hydrolyzed Vegetable Protein Made by Partial Hydrolysis of Rice Bran Protein ARPORN JARUNRATTANASRI,† CHOCKCHAI THEERAKULKAIT,*,† AND KEITH R. CADWALLADER‡ Department of Food Science and Technology, Kasetsart University, Bangkok, 10900, Thailand, and Department of Food Science and Human Nutrition, University of Illinois, 1302 West Pennsylvania Avenue, Urbana, Illinois 61801

Hydrolyzed vegetable protein (HVP) was prepared from rice bran protein concentrate (RBPc) by partial hydrolysis with aqueous 0.5 N HCl at 95 °C for 12 or 36 h (H-RBPc-12 and H-RBPc-36, respectively). Aroma components of the RBPc and the HVPs were characterized by gas chromatography-olfactometry, gas chromatography-mass spectrometry, aroma extract dilution analysis, and calculation of odor activity values (OAVs). The predominant odorants in RBPc were 3-methylbutanal, hexanal, 2-aminoacetophenone, (E)-2-nonenal, phenylacetaldehyde, and β-damascenone. Among these, the odor of 2-aminoacetophenone, present at 59 ng/g in RBPc, was reminiscent of the typical odor of RBPc. Most of the predominant odorants had higher log3FD factors in the H-RBPc36 as compared to H-RBPc-12. Aroma impact compounds of H-RBPc-12 and H-RBPc-36 were 2-methoxyphenol (guaiacol), 4-hydroxy-2,5-dimethyl-3(2H)furanone, 3-hydroxy-4,5-dimethyl-2(5H)furanone (sotolon), vanillin, 3-methylbutanal, (E)-2-nonenal, 4-vinyl-2-methoxyphenol (p-vinylguaiacol), and β-damascenone. Guaiacol had the highest OAV values of 2770 and 17650 in H-RBPc-12 and H-RBPc-36, respectively. KEYWORDS: Acid hydrolysis; rice bran protein concentrate; hydrolyzed vegetable protein; aroma; flavor

INTRODUCTION

Rice bran (RB) is the outer brown layer, including the rice germ that is removed during the milling of brown rice. Fujimaki (1) reported that although a large quantity of RB is available throughout the world, little is actually consumed by humans because of its unpleasant flavor. Over 270 aroma compounds were detected in RB, of which approximately 170 compounds were identified (1, 2). These included minor amounts of p-vinylguaiacol and p-vinylphenol that were reported to impart unpleasant medicinal odors. The number and complexity of compounds that contribute to the flavor of various protein sources are considerable. The biggest contribution to flavor is made by free amino acids and peptides. The most common reaction causing protein fragmentation is hydrolysis (3), which yields peptides and free amino acids. These compounds can influence perceived flavor or can serve as flavor precursors, such as in the Strecker degradation of amino acids. In addition, amino acids and peptides may undergo many transformations, such as reaction with reducing sugars (Maillard reaction) to form aroma compounds and pigments (4). Chemical hydrolysis remains one of the most popular forms of protein modification (5). Peptide bond hy* To whom correspondence should be addressed. Tel: +66-2-502-5032. Fax: +66-2-562-5020. E-mail: [email protected]. † Kasetsart University. ‡ University of Illinois.

drolysis, by either a strong acid or a base, may be used to yield smaller peptides with a more uniform molecular size or free amino acids. Hydrolyzed vegetable protein (HVP) produced commercially by acid hydrolysis has been used to produce various types of flavors by the Maillard reaction. Enzymehydrolyzed vegetable protein (EVP), where the protein source is hydrolyzed by proteolytic enzymes, is an alternative to HVP. However, the acidic hydrolysate is dark brown with a strong savory flavor, whereas the enzymatic hydrolysate is usually lighter in color and has a much less pronounced meaty or savory flavor (3). Hydrochloric acid (HCl) is commonly used in the production of protein hydrolysates because it works quickly and yields a fully hydrolyzed product with a highly acceptable savory profile (5). Currently, the formation of potentially carcinogenic chlorohydrins (e.g., 3-chloro-1,2-propanediol and 1,3-dichloro2-propanol) in HVP from the use of concentrated HCl for hydrolysis has been a serious concern in many countries (6). Collier et al. (6) reported 0.9-5.7% of total chloropropandiols by weight of substrate in HVP produced under strong hydrolysis conditions (5.5 M HCl, 107 °C). To avoid the formation of chlorohydrins, mild hydrolysis reactions such as the use of proteolytic enzyme (3) or mild acid hydrolysis conditions (7) should be used. The use of mild acid hydrolysis conditions (low concentration of HCl and low temperature) leads to partial hydrolysis, which generates a mixture of free amino acids and small and large peptides. The size of the peptide, the position

10.1021/jf0631474 CCC: $37.00 © 2007 American Chemical Society Published on Web 03/17/2007

J. Agric. Food Chem., Vol. 55, No. 8, 2007

Aroma of Acid-Hydrolyzed Rice Bran Protein of the amino acids within the peptide, and the resistance of the peptide bond to further hydrolysis are important in flavor formation (8). Ho et al. (9) studied amino acids and peptides as flavor precursors in a model Maillard reaction by heating amino acids or peptides separately with glucose at 180 °C under pH 4-5 for 2 h. The reaction of glucose with glycine or triglycine produced more pyrazines than when glucose was reacted with diglycine or tetraglycine. One of the most important flavor precursor components in defatted rice bran (DRB) is protein. DRB may be used as a raw material for the production of HVP (10). Hamada (11) demonstrated that peptides, which are generated from DRB by enzymatic hydrolysis, contained substantial amounts of glutamic acid and could serve as a flavor-enhancing ingredient after further deamination. Jarunrattanasri et al. (12) also produced HVP from DRB by first preparing a rice bran protein concentrate (RBPc) prior to hydrolysis with 4.0 N HCl at 95 °C for 72 h. However, HVP prepared from RBPc by acid hydrolysis using mild reaction conditions has not been previously reported. Use of a mild acid hydrolysis of DRB protein concentrate may add value to underutilized RB through its conversion, via acid hydrolysis of protein and the subsequent Maillard reaction of amino acids and peptides with carbonyl compounds in RBPc, to an aroma or flavoring ingredient. In the analysis of flavor, gas chromatography-mass spectrometry (GC-MS) can be used to selectively analyze aroma compounds once their spectral and chromatographic properties are known. However, the task of determining which compounds in a sample are odor-active requires a bioassay. Teranishi (13) concluded that the constituent(s) that contribute to the characteristic sensory properties of the food product should first be investigated. Gas chromatography-olfactometry (GCO) is a method that reveals odorants in terms of their pattern of smell activity, thus eliminating odorless compounds from consideration. The purpose of our study was to identify and quantify the predominant aroma components that are generated by partial acid hydrolysis of RBPc. MATERIALS AND METHODS Chemicals. Analytical-grade reference compounds were obtained from Aldrich Chemical Co. (St. Louis, MO) except for (E)-βdamascenone, which was provided by Firmenich Co. (Princeton, NJ). 2-Acetyl-1-pyrroline was a gift from Dr. R. Buttery (Agricultural Research Service, U.S. Department of Agriculture, WRRC, Albany, CA). trans-4,5-Epoxy-(E)-2-decenal was synthesized using a previously published procedure (14). (Z)-2-Nonenal was synthesized from (Z)-2nonen-1-ol (Bedoukian Research Inc., Danbury, CT) by oxidation with Dess-Martin periodinane (0.3 M in dichloromethane; Aldrich Chemical Co.) using a published procedure (15). Odorless distilled water was prepared by boiling glass-distilled water in an open flask until its volume was reduced by one-third of the original volume. DRB was prepared from freshly milled RB from jasmine rice (Oryza satiVa, Variety Khao Dawk Mali 105) by defatting twice with hexane (RB to hexane at a ratio of 1:3, w/v), followed by air drying overnight in a fume hood, grinding in a sample mill, and sieving through an 80 mesh screen (16). DRB was packed in polyethylene bags and stored at 5 °C until needed. The alkali extraction procedure followed by isoelectric precipitation was used to prepare RBPc from DRB (17). Hydrolysis. RBPc (containing 100 g of protein) plus 500 mL of aqueous 0.5 N HCl was placed in a 1 L amber glass bottle. After purging with purified nitrogen gas for 10 min, the bottle was sealed with a PTFE-linned cap and then incubated for either 12 or 36 h at 95 °C. The hydrolysate was adjusted to pH 6.0 with aqeuous 1.0 N NaOH, then filtered through Celite (0.1% w/v added to hydrolysate prior to filtration), freeze-dried, and stored at -5 °C until analysis (18). The degree of hydrolysis (DH) was defined as the percentage of peptide

3045

bonds cleaved (19). DH values were determined by analyzing free and total amino acids (19) and calculated as follows (20):

%DH )

number of free amino groups × 100 total number of amino acid residues

Proximate Analysis. The percent moisture (method 935.29), crude fat (method 945.16), crude protein (method 990.03), and ash (method 923.03) of samples were determined according to AOAC procedures (21). Carbohydrate was determined by difference (22). Determination of Free and Total Amino Acids. Free and total amino acids were determined using a Beckman 6380 Amino Acid Analyzer (Beckman Instruments Inc., Palo Alto, CA) with a 10 cm ion exchange column and lithium citrate buffer supplied by Beckman. Detection was via postcolumn derivatization using ninhydrin. Calibration was achieved using mixed external free amino acid standards. The analytical conditions were the same as previously published (12). Isolation of Volatile Compounds. RBPc or HVP (20 g) was dissolved in 100 mL of odorless distilled water and spiked with 25 µL of internal standard solution (5.00 µg/µL of 2-ethyl butyric acid in methanol as acidic fraction internal standard and 4.31 µg/µL of 2-methyl-3-heptanone in methanol as neutral/basic fraction internal standard). The solution was extracted (with shaking for 30 min) two times with 50 mL of diethyl ether. The solvent extract was subjected to a high vacuum distillation (approximately 5 × 10-5 Torr operating vacuum level) cleanup step as previously described (23) for 3 h to further remove nonvolatile residue, with the sample kept at room temperature for the first 1 h and then warmed to 45 °C using a water bath. The solvent extract was evaporated to 20-25 mL using a Vigreux column in a 40 °C water bath. To separate the acidic volatiles from the neutral/basic volatiles, the extract was washed three times with aqueous 5% Na2CO3 solution (3 × 20 mL), and the organic layer containing the neutral/basic volatiles was collected. The aqueous layer was then acidified to pH 3 with 10% aqueous HCl and extracted with diethyl ether (3 × 15 mL). The organic layer containing the acidic volatiles was collected. Each fraction was then concentrated under a gentle stream of nitrogen gas to 10 mL, dried over 2 g of anhydrous sodium sulfate, and further concentrated to 250 and 500 µL under a nitrogen gas stream for neutral/basic and acidic fractions, respectively. Samples were prepared in duplicate and kept at -70 °C until analysis. GC-MS. GC-MS was performed on a 6890 GC /5973 mass selective detector (MSD) (Agilent Technologies, Inc., Palo Alto, CA). One microliter of each extract was injected using cool on-column mode (38 °C initial temperature, with +3 °C oven temperature tracking) into a DB-FFAP (30 m × 0.25 mm i.d. × 0.25 µm film; J&W Scientific, Folsom, CA) or HP5-MS (30 m × 0.25 mm i.d. × 0.5 µm film; Agilent Technologies, Inc.) fused silica capillary column. The GC oven temperature was programmed from 35 to 225 °C at rate of 4 °C/min with initial and final hold times of 5 and 30 min, respectively. Helium was used as the carrier gas at a constant flow of 1 mL/min. The mass spectrometer was operated in electron impact mode with an electron energy of 70 eV and an emission current of 50 mA. The mass spectrometer scanned from m/z 29 to m/z 400 at 1.9 scans/s. The MSD interface temperature was 280 °C. GCO. GCO was performed on an Agilent 6890 GC equipped with a flame ionization detector (FID) and olfactory detection port (OPD2, Gerstel, Germany). Two microliters of each was injected using a direct on-column mode (38 °C initial temperature, with +3 °C oven temperature tracking) into a DB5-MS or DB-FFAP (15 m × 0.32 mm i.d. × 0.5 µm film thickness, J&W Scientific) fused silica capillary column. The oven was held at 35 °C for 5 min, ramped at 10 °C/min to 225 °C, and then held for 10 min. The column effluent was spilt 1:1 between the FID and an olfactory port using deactivated fused silica tubing (1 m × 0.15 mm i.d). Helium was used as the carrier gas at a constant flow of 2.2 mL/min. The aroma-active compounds were evaluated by aroma extract dilution analysis (AEDA) (24) on 1:3 dilution series of the aroma extracts. The ratio of the concentration of the odorant in the initial extract to its concentration in the most dilute extract in which the odor was detected by GCO was the flavor dilution

J. Agric. Food Chem., Vol. 55, No. 8, 2007

3046

Jarunrattanasri et al.

Table 1. Proximate Compositiona (% by Weight) of RB, DRB, and

RBPc sample

moisture

crude fat

protein

carbohydrate (by difference)

ash

RB DRB RBPc

8.90 b 12.09 a 5.02 c

19.54 b 5.26 c 22.35 a

12.35 c 13.97 b 53.78 a

51.34 a 60.10 a 16.61 b

7.57 b 8.58 a 2.24 c

a Means

(n ) 3) in the same column with different letters are significantly different

(p e 0.05).

Table 2. Protein Contenta and Percent DHa of RBPc and Acid-Hydrolyzed RBPcs Prepared at 12 (H-RBPc-12) and 36 h (H-RBPc-36) sample

% protein (wet basis)

% DH

RBPc H-RBPc-12 H-RBPc-36

53.8 a 23.4 c 23.9 b

13.3 b 24.0 a

a Means (n ) 3) in the same column with different letters are significantly different (p e 0.05).

(FD) factor (24). GCO was then performed by two experienced analysts. Results of AEDA are reported as average log3FD factors (25). Compound Identification. Compounds were positively identified by comparing their mass spectra and retention indices (RI) on both DB-FFAP and HP5-MS columns to those of authentic reference compounds analyzed under identical conditions. When mass spectra were unavailable, compounds were tentatively identified by comparison of their RI values and odor properties on both DB-FFAP and HP5-MS columns to those of authentic reference compounds. RI values were determined by analyzing a series of n-alkanes (C6-C28 or C6-C18 for DB-FFAP or HP5-MS, respectively) as described by van Den Dool and Kratz (26). Compound Quantification. The concentration of an aroma component was calculated by internal standard methodology by GC-MS using a DB-FFAP column as described above. Prior to GC-MS, reference standards (at three different levels) plus internal standards were diluted in deodorized water and subjected to solvent extraction, high vacuum distillation, and compound class fractionation. Because water was used as the matrix and not mimic matrices based on RBPc or the hydrolysates, the values reported herein are considered relative concentrations. The relative concentration of a compound was calculated as previously described by Zhou et al. (25). Statistical Analyses. Three sample replications were performed. Data were analyzed by least significant differences procedures to separate means, and differences were reported as significant at p ) 0.05 (27). RESULTS AND DISCUSSION

Chemical Composition of RB and RBPc. Protein contents of full fat RB and its protein concentrate (RBPc) were 12.35 and 53.78%, respectively (Table 1). Crude fat increased slightly from 19.54 to 22.35% in RB and RBPc, respectively (Table 1), which could be explained by the retention of residual protein-lipid complexes in RBPc that were not fully extracted in the defatting step. Cheftel et al. (28) pointed out that lipid oxidation followed by protein-lipid covalent interactions can take place in some foods and feeds. The DH of acid-hydrolyzed RBPcs as a function of hydrolysis time is shown in Table 2. Free and total amino acid contents (% dry basis) of RBPc and free amino acids of HVPs prepared with 0.5 N HCl for 12 (HRBPc-12) and 36 h (H-RBPc-36) are given in Table 3. RBPc was obtained by isoelectric precipitation at pH 4.5; therefore, in RBPc, the major protein is acidic. This agrees with the results of free and total amino acid analysis in that glutamic acid was

Table 3. Free and Total Amino Acids of RBPc and Free Amino Acid of Acid-Hydrolyzed RBPcs Prepared at 12 (H-RBPc-12) and 36 h (H-RBPc-36) free amino acid (% dry basis)a

amino acid

MW

RBPc

H-RBPc-12

H-RBPc-36

total amino acids of RBPc (% dry basis)

alanine arginine aspartic glutamic glycine histidine isoleucine leucine lysine methionine phenylalanine proline serine threonine tyrosine valine

89.09 174.20 133.10 147.13 75.07 155.15 131.18 131.18 146.19 149.21 165.19 115.13 105.09 119.12 181.19 117.13

0.010 c NDb 0.020 c 0.034 c 0.005 c ND 0.006 c ND 0.010 c 0.002 b ND ND 0.005 c 0.004 b ND ND

0.327 b 0.751 a 1.390 b 0.494 b 0.446 b 0.072 b 0.048 b 0.138 b 0.101 b 0.066 b 0.083 b ND 0.156 b 0.065 b 0.069 b ND

0.865 a 0.807 a 2.201 a 3.989 a 0.879 a 0.172 a 0.203 a 0.471 a 0.279 a 0.246 a 0.236 a 0.452 a 0.438 a 0.241 a 0.251 a 0.070 a

4.095 5.913 4.865 8.840 3.531 1.932 1.948 4.699 2.830 1.114 2.564 2.595 3.185 2.268 2.344 3.131

a Means (n ) 3) in the same row with different letters are significantly different (p e 0.05). b ND, not detected.

in highest abundance in RBPc (Table 3). Juliano (29) and Wang et al. (16) also reported that glutamic acid was most abundant among the total amino acids in RB. Therefore, glutamate residues should provide the main acidic character to the protein in RBPc. Glutamic acid also was in highest abundance of free amino acid in H-RBPc-36. Jarunrattanasri et al. (12) also found glutamic acid in highest abundance in HVP prepared from RBPc by hydrolysis with 4.0 N HCl at 95 °C for 72 h. On the other hand, aspartic acid was the predominant free amino acid in H-RBPc-12. The above results are reasonable since glutamic acid, arginine, and aspartic acid are the major amino acids among the total amino acids in RBPc. Aroma-Active Compounds. Aroma compounds of intermediate and low volatility were isolated by liquid-liquid solvent extraction, followed by a high vacuum distillation cleanup step. A total of 27 odorants were detected by GCO and AEDA. Seven, 19, and 21 volatile compounds with log3FD factors g 2 were detected in RBPc, H-RBPc-12, and H-RBPc-36, respectively (Table 4). Previously, 30 odorants were reported in HVP produced from RB protein made by hydrolysis with concentrated (4 N) HCl (12). Twenty-one of those compounds were also detected in the present study in H-RBPc-12 and H-RBPc-36. Most of the predominant odorants identified in the present study had higher log3FD factors in H-RBPc-36 as compared with H-RBPc-12. (E)-2-Nonenal (no. 28) and trans-4,5-epoxy-(E)2-decenal (no. 33) were an exception and had higher log3FD factors in H-RBPc-12 than in H-RBPc-36. Results showed that odorants originated from lipid oxidation, lignin degradation, and Maillard reactions as discussed below. Acidic Fraction. Among the acidic odorants, hexanoic acid (no. 8) and vanillin (no. 18) had the highest log3FD factors in RBPc and imparted sweaty and vanilla-like notes, respectively. Vanillin was also a predominant odorant of both HVPs. In a previous study, vanillin was not reported as a predominant odorant in HVP produced from RBPc by hydrolysis with concentrated (4 N) HCl (12). Other acidic odorants found with high FD factors in H-RBPc-12 and H-RBPc-36 were 2-methoxyphenol (guaiacol, no. 9), 4-hydroxy-2,5-dimethyl-3(2H)-

J. Agric. Food Chem., Vol. 55, No. 8, 2007

Aroma of Acid-Hydrolyzed Rice Bran Protein

3047

Table 4. Potent Odorants in RBPc and Acid-Hydrolyzed RBPcs Prepared at 12 (H-RBPc-12) and 36 h (H-RBPc-36) RIb no.a

odorants

3 6 8 9 11 12 13 14 15 16 17 18

acetic acidd 3-methylbutyric acidd hexanoic acidd 2-methoxyphenol (guaiacol)d 4-hydroxy-2,5-dimethyl-3(2H)furanonef octanoic acidd 4-methylphenol (p-cresol)d 4-vinyl-2-methoxyphenol (p-vinylguaiacol)d 3-hydroxy-4,5-dimethyl-2(5H)furanone (sotolon)f 2,6-dimethoxyphenol (syringol)d phenylacetic acidd vanillind

acidic fraction vinegar sweaty sweaty smoky burnt sugar sweaty stable, dung clove seasoning smoky honey vanilla

19 20 22 23 24 25 26 27 28 29 30 31 32 33 34

2-methylbutanald 3-methylbutanald hexanald 1-octen-3-onef 2-acetyl-1-pyrrolined dimethyl trisulfidef 3-(methylthio)propanal (methional)d (Z)-2-nonenalf (E)-2-nonenald (E,Z)-2,6-nonadienalf phenylacetaldehyded β-damascenonef 2-phenylethanold trans-4,5-epoxy-(E)-2-decenalf 2-aminoacetophenoned

odor

Log3FDc

FFAP

DB5

RBPc

H-RBPc-12

H-RBPc-36

1449 1667 1839 1867 2028 2046 2089 2175 2198 2273 2565 2571

−e 888 − 1096 1097 − 1098 1365 1105 1367 1292 1415

− 1 2 1 − − − − − − − 2

2 2 1 4 4 − 2 7 4 2 3 4

3 1 1 5 6 2 3 7 6 3 4 6

neutral/basic fraction malty 915 malty 922 grass 1087 mushroom 1303 cooked rice 1340 cabbage 1381 cooked potato 1455 fatty, hay 1513 lipstick, waxy 1543 cucumber 1596 rosy 1652 applesauce 1830 floral 1920 unripe, metallic 2017 tortilla 2225

659 649 799 979 922 972 906 − 1171 1156 1050 1384 1139 1389 1315

1 1 2 1 1 1 1 1 3 1 3 3 − − 4

1 1 1 3 3 − 3 3 4 1 1 6 3 5 2

2 2 1 3 3 1 5 1 3 3 1 7 3 3 2

a Numbers correspond to those Table 5. b Retention index calculated from GCO data. c Average Log FD (n ) 2) determined on DB-FFAP column. d Compound positively 3 identified by comparison of its RI values, odor properties, and mass spectrum with reference compound. e Not available. f Compound tentatively identified by comparison of its RI values and odor properties with reference compound.

furanone (no. 11), 3-hydroxy-4,5-dimethyl-2(5H)furanone (sotolon, no. 15), and phenylacetic acid (no. 17). Both guaiacol (no. 9) and 2,6-dimethoxyphenol (syringol, no. 16) imparted strong and potentially undesirable smoky notes to both HVPs. 4-Vinyl-2-methoxyphenol (p-vinylguaiacol, no. 14) contributed a spicy, clovelike note at high log3FD factors of 7 in both of HVPs, but it was not detected in RBPc. Phenols and guaiacols are the major contributors to wood smoke aroma (30). The presence of guaiacol has been previously reported in cooked brown rice (31), in both acid- and enzyme-hydrolyzed soy protein (32). and in soybean-based enzyme hydrolysates (33). Maga (34) pointed out that in wood smoke phenol and its derivatives are formed as a result of lignin degradation. 4-Hydroxy-2,5-dimethyl-3(2H)furanone (no. 11) imparted a burnt sugar note at high log3FD factors of 4 and 6 in H-RBPc12 and H-RBPc-36, respectively. While this compound is a predominant odorant in both HVPs, it was less important in an HVP made using concentrated (4 N) HCl (12). Schieberle (35) stated that 4-hydroxy-2,5-dimethyl-3(2H)furanone will usually be formed in some heat-processed foods via the Maillard reaction, and it has been detected in several foods, e.g., roasted almond, wheat bread crust, and popcorn. In particular, hexoses can be rapidly converted into 4-hydroxy-2,5-dimethyl-3(2H)furanone (35). This is supported by Toledo et al. (36) who demonstrated that 4-hydroxy-2,5-dimethyl-3(2H)furanone was formed only from arginine and rhamnose under acidic conditions in a model citrus juice system, whereas at pH 6-8, 4-hydroxy2,5-dimethyl-3(2H)furanone was formed from glucose and fructose. 4-Hydroxy-2,5-dimethyl-3(2H)furanone was also formed from rhamnose and alanine at pH 3.5 after refluxing for 5 h (37).

Sotolon (no. 15) with high log3FD factors of 4 and 6 in H-RBPc-12 and H-RBPc-36, respectively, was responsible for a seasoning or currylike odor. This compound was reported as an important odorant of an HVP made by hydrolysis of RBPc with concentrated (4 N) HCl (12). Sotolon was previously reported to contribute to the burnt/sweet aroma of cane sugar (38). Similar to compounds imparting smoky notes, some other odorants such as 3-methylbutyric acid (no. 6), hexanoic acid (no. 8), and octanoic acid (no. 12) may be considered undesirable because they imparted sweaty odors. Neutral/Basic Fraction. Predominant neutral/basic odorants in RBPc were (E)-2-nonenal (no. 28), phenylacetaldehyde (no. 30), β-damascenone (no. 31), and 2-aminoacetophenone (no. 34). 2-Aminoacetophenone may be a characteristic aroma component of RBPc, since not only did it have the highest log3FD factor in RBPc but its odor is very similar to that of RBPc. 2-Aminoacetophenone has been previously reported as a volatile constituent of scented rice (31), and it contributes to the offflavor of dried milk (39) and renet casein powders (40). Buttery and Ling (41) suggested that 2-aminoacetophenone may be formed by breakdown of tryptophan during lime treatment of corn during the preparation of masa dough. It has been clearly demonstrated that the most likely pathway of formation of 2-aminoacetophenone is from the degradation of tryptophan and/ or indole-3-acetic acid, which are present in wine musts (cited by ref 42). In the present study, tryptophan was not detected among the free or total amino acids in Table 3 since it is present in only trace amounts as a free amino acid in RBPc (43); furthermore, tryptophan is readily degraded during the acid hydrolysis step (44).

3048

J. Agric. Food Chem., Vol. 55, No. 8, 2007

Jarunrattanasri et al.

Table 5. Concentrations and Odor Activity Values (OAVs) of Selected Volatile Components of RBPc and of Acid-Hydrolyzed RBPcs Prepared at 12 (H-RBPc-12) and 36 h (H-RBPc-36) RIb

no.a 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

compound

FFAP

concentration (ng/g)d

HP5

threshold (ng/g in water)c

RBPc

OAVe

H-RBPc-12

H-RBPc-36

RBPc

H-RBPc-12

H-RBPc-36

9537 (674) B 432 (30) 3516 (248) B 76 (5.6) B 86 (6.4) B 247 (18) 188 (13) B 1092 (77) B 6934 (490) B

10657 (753) B − 4501 (318) A 126 (8.0) A 132 (9.0) A − 225 (16) B 1306 (93) B 44130 (3120) B

5.0 −